Snacking has become a popular habit among children and teenagers (1). At the same time, overweight and obesity have reached huge proportions, affecting young individuals, particularly in Italy, where the childhood obesity is higher than in most European countries (2). Snacking has been considered one of the main contributors to overweight because of the increased consumption of energy-dense, high-sugar, high-fat foods (3–8). Other studies have, however, failed to establish a correlation between snacking and overweight (9–14). Such variability could be explained by the definition variability of “snack” (13). The term “snack” generally designates a small number of items characterized by low nutrient and high energy density (eg, candies, cookies, salty snacks, soda). An alternative definition is based on the time criterion, leading us to consider snacks as the eating occasions between the main meals (13). From a nutritional perspective, varying the composition of snacks with healthier food variety has proved to be a successful strategy in overweight children who are trying to lose weight (15–16). According to Drummond et al (17), snacking per se may not necessarily predispose an individual to become overweight because a higher eating frequency may be favorable in terms of body weight control and energy balance, lowering caloric intake when replacing the main meals. In relation to meal frequency, some authors recognize an inverse relation between the number of eating occasions and the prevalence of overweight (11,14,18–19). Finally, the time of snacking may affect metabolism and energy expenditure differently. This is because diet-induced thermogenesis is reduced when a test snack is consumed in the evening compared with both morning and afternoon (20), and nighttime snacking decreases whole-body fat oxidation (21). This seems to be important, considering the modern trend of eating less in the morning and more during the day in children and adolescents (22). Even if underreporting in food questionnaires could affect many of these findings, the debate is still open, and snacking deserves more research because it could help in developing strategies to control or reduce body weight in pediatric subjects.
The purpose of this study was to evaluate the relation between the energy content, frequency, and timing of snacking, and overweight/obesity in 11- to 13-year-old Italian adolescents.
All of the students attending the first year of middle school in Torino (Italy) routinely underwent a health status and physical performance assessment conducted by the Sports Medicine Institute (“Bambini a Torino” project). In academic year 2011–2012, 7263 students attended the first year of middle school in Torino. Among them, 6876 participated in the assessment with participation rate of 94.7% (6876/7263).
A week before the assessment, students were administered questionnaires on health status, exercise levels, present and past illness, and medications by schoolteachers, and were asked to have them filled out by their parents at home. In 400 students, selected by a simple random sampling, a dietary questionnaire was added to the health status questionnaire. The present study was based on these 400 students.
Children's parents or legal representatives signed a consent form for the study participation. The study protocol was approved by the local ethical committee, and the procedures were in compliance with the Declaration of Helsinki principles.
Weight and height were measured for all of the participants. Weight was measured to the nearest 0.1 kg, and height was measured to the nearest 0.1 cm using a stadiometer (SECA Model 711; SECA, Hamburg, Germany), with the participants wearing light clothes and no shoes. Body mass index (BMI) was defined as body mass divided by height in square meters (kg/m2), and the BMI percentile values were obtained from the Italian growth charts, referring to northern/central Italy percentiles (23). Subjects who had a BMI ≥85th percentile were classified as overweight/obese.
Data related to the frequency, duration of each session, and intensity of exercise performed by the participants outside school were collected by the questionnaires administered to the parents. Exercise was considered as any planned, structured, and repetitive physical activity performed outside gymnastics classes at school. The mean weekly exercise level was calculated as the product of duration and frequency of each activity (in hours per week), weighted by an estimate of the metabolic equivalent of the activity (MET), and summed for the activities performed (24). The whole cohort was divided into tertiles of exercise, and individuals were classified as sedentary, moderately active, and active. The cutoff points in the whole cohort were ≤10, 11 to 29, and ≥30 MET hours per week, respectively, and were similar to those found in another European cohort of adolescents (25).
The parents of the 400 randomly selected subjects were asked to complete both a validated 24-hour food recall questionnaire (26) and a 19-item food-frequency grid with an instruction sheet. In the food-recall questionnaire, different serving sizes were specified for each item, using commonly used portions; furthermore, the use of measuring guides (cups, spoons, glasses) aided respondents in estimating the amounts of foods consumed. The temporal pattern of food consumption was recorded, because parents were asked to list foods eaten at each meal, including snacks.
The grid evaluated the mean weekly frequency of consumption of 19 items (eg, biscuits, fruit juices, soft drinks, fruits, savory snacks, pizzas, crackers, potato chips, popcorn, candies, bread, vegetables, cold cuts, sweets, milk/yogurt) and was developed using a validated semiquantitative food-frequency questionnaire (27). The parents indicated how many times per week their child consumed each food included in the list by ticking 1 of 3 boxes referring to different ranges of consumption frequency, which varied in accordance to the kind of food. The data obtained from the grid were separately analyzed and used to check the data collected by the food-recall questionnaire. No missing cases occurred. Of the 7263 students attending the first year of middle school in Torino, 3452 (47.5%) ate in the school cafeteria. Similarly, 184/400 (46.0%) participants of our random sample ate at the cafeteria. The parents were given the cafeteria menus, so that they knew types and portions of the distributed food.
During the visit at the Sports Medicine Institute, a skilled dietitian interviewed the participants in case of uncertain answers or discrepancy between data from the grid and the food-recall questionnaire, asking them to indicate how often per day and how many days per week they had usually consumed a specific food and its serving size with the aid of sample photos; 155/400 (38.8%) participants were interviewed. There was no significant difference in age, sex distribution, ethnicity, and BMI between the interviewed and noninterviewed participants.
The food-recall questionnaire data were analyzed using the software Win Food Pro 3 (Medimatica, Colonnella, Teramo, Italy). The estimated basal metabolic rate (BMR) for 10- to 18-year-old individuals was calculated using the following formula in accordance with Schofield (28): for boys, BMR = 0.068 weight + 0.574 height + 2.157; for girls, BMR = 0.035 weight + 1.948 height + 0.837. Then, the estimated energy requirement was calculated by multiplying the BMR by a coefficient corresponding to 1 of the 3 exercise levels (sedentary, moderately active, and active), in line with the Italian and European guidelines (29). Finally, the reliability of the reported energy intake was assessed by calculating the ratio of estimated energy intake to predicted BMR; subjects with a ratio <0.88 and >2.72 were classified as underreporters or overreporters, respectively (30); 12 underreporters and 2 overreporters were identified.
Diet quality was evaluated with the Simple Dietary Quality Index (SDQI) score, first used in Denmark, comparing saturated fatty acids (SFAs); and fiber intake to the Danish dietetic recommendations (31). An adjustment in the dietary fiber estimation was needed for our cohort because the Italian guidelines recommended 2 g/MJ for children/adolescents. For each individual, a score was calculated for SFA and dietary fiber, and the minimum and maximum scores for each nutrient were 0 and 100, respectively. The SDQI score was the mean of the scores for SFA and fiber; the higher the score, the healthier the diet (31).
Snacking episodes were defined as eating occasions not included in breakfast, lunch, and dinner, in accord with the literature (8). The snacking occasions were reported by the participants’ parents, along with a detailed description of the kind, amount, and time of the snacks. Morning snacking was considered as food consumption between breakfast and lunch, afternoon snacking between lunch and dinner, and evening snacking after dinner. Most participants consumed snacks at different occasions during the day. Individuals were categorized as predominantly morning, afternoon, and evening snackers if the highest caloric intake from snacks derived from the morning, afternoon, and evening snacks, respectively. Furthermore, analyses were repeated after considering snackers whose daily caloric intake from snacks was ≥15% of the total daily caloric intake (13).
Based on available data (1,6), a sample size of 371 patients had a statistical power of 80% to detect a 2-fold higher prevalence of overweight/obesity in individuals consuming ≥3 snacks/day, with a 2-tailed α-value = 0.05. Normality distribution of data was checked by the Kolmogorov-Smirnov normality test. Exercise level, expressed as MET hours per week, did not show a normality distribution, and differences among groups for this variable were analyzed by the Kruskal-Wallis test. In addition, values were log-transformed, thereby obtaining a normal distribution, and the log-transformed values were used in the regression model (see below).
The analysis of variance and the χ2 test were performed to assess the differences in the continuous and categorical variables, respectively. A Poisson regression analysis with robust variance was used to estimate the relative risk of prevalent overweight/obesity by energy content, number, and timing of consuming the most energy-dense snack, after adjusting for age (continuous variable), sex (boys = 1), exercise levels (log-MET hours per week), the SDQI score (continuous variable), and total caloric intake (continuous variable). The same analyses were replicated in the subgroup of participants who consumed ≥15% of the total daily caloric intake as snacks (n = 272).
Table 1 summarizes the characteristics of the whole cohort of adolescents and selected sample undergoing the medical assessment. There was no difference between the adolescent cohort as a whole and the randomly selected group.
On average, participants consumed 1.9 ± 0.7 snacks per day (boys 1.9 ± 0.8, girls 2.0 ± 0.7), contributing to 18.1% of the daily energy intake. The most frequently consumed snacks were sweets, savory snacks, milk/yogurt, fruit juice, and soft drinks; occasionally fruits, bread, cereals and grains; and rarely fresh vegetables and cold cuts.
The overall prevalence of overweight/obesity was 16.5% (66/400), and 29.5% of participants performed no exercise other than gymnastic classes at school. The prevalence of overweight/obesity did not differ among individuals who ate at the cafeteria (16.4%) and those who did not (16.6%).
Participants were divided according to the percentage of daily energy derived from snacks (Table 2). A low number of subjects (13/400, 3.3%) did not consume any snacks and showed lower total caloric intake, lower exercise levels, and a significantly higher prevalence of overweight/obesity. Notably, 6 of the 12 underreporters were found in this category. To avoid the possibility that the association between no snacking and BMI may be the result of reverse causality (ie, subjects did not consume snacks because they were trying to lose weight), these individuals were excluded, and data from 387 participants were analyzed. Increasing contribution of snacks to daily energy intake, from <15% to >20%, was associated with a significant increase in total caloric intake and a higher prevalence of overweight.
The number of snacks consumed during the day ranged from 1 to 5. The snack number (Table 3) was positively correlated with higher total caloric intake, reduced MET hours per week, and increased overweight/obesity prevalence, varying from 9.6% to 22.6%. The increased caloric intake, however, was characterized by higher fiber and lower SFA intake.
Participants were also classified according to the timing of consuming the most caloric snack of the day (Table 4). In most cases, the morning (mean percentage of the total snack calories 63.9%) and the afternoon (64.4% of the snack calories) snacks showed the highest caloric content, whereas the evening snack (69.3% of the snack calories) was more calorie dense in 28 of 387 cases only. The percentage of energy taken at breakfast, lunch, and dinner did not differ among morning, afternoon, and evening snackers (data not shown). The evening snackers showed increased fiber intake and a better SDQI score, but their overweight prevalence was high (28.6%), and in 39% of these individuals the evening snack was the only snack consumed during the day. Finally, the characteristics of the participants were evaluated by overweight/obesity status (Table 5): individuals with BMI ≥85th percentile showed significantly higher SFA intake and lower fiber and SDQI score. In a Poisson regression model, the variables associated with BMI ≥85th percentile were fiber intake (relative risk [RR] 0.88, 95% confidence interval [CI] 0.83–0.93, P < 0.001), SFA intake (RR 1.12, 95% CI 1.04–1.22, P = 0.004), and the SDQI score, which is a combination of these 2 variables (RR 0.95, 95% CI 0.93–0.96, P < 0.001). In the same model, increased snacking calories and frequency and large evening snacks were significantly associated with BMI ≥85th percentile (Table 6). Analyses were repeated after considering only snackers who consumed ≥15% total energy intake as snacks and after excluding the 6 underreporters and the 2 overreporters; the results did not change. Finally, frequency and timing of snacking were analyzed together in the same model, and both of them showed a positive association with overweight/obesity (RR 1.73, 95% CI 1.09–2.75, P = 0.02 for consuming ≥3 snacks per day; RR 1.98, 95% CI 1.18–3.33, P = 0.01 for consuming the highest-calorie snack at evening).
The calories, number, and timing of snacks are associated with overweight/obesity, independent from diet quality and exercise frequency in a sample of 11- to 13-year-old Italian adolescents. Snacking is a common habit in adolescents: 97% of our subjects usually consumed at least 1 snack during the day, as in previous reports (9,14,32). Snacks may represent a good source of fiber. If fruits and fresh vegetables were consumed daily as snacks, it would be easier to achieve the recommended nutritional goals (8,32,33); however, snacks generally fail to improve diet quality because they are usually energy-dense and high-carbohydrate foods, despite that they have a lower fat content than the main meals (8,11,32,33). In our sample, adolescents with higher snacking frequency or quantity showed significantly higher fiber and total energy intake and lower SFA intake. Therefore, dietary quality, as evaluated by the SDQI score, was not lower in those who consumed the largest amount of calories as snacks (“big snackers”).
In our study, the prevalence of overweight/obesity was higher in participants consuming >20% calories from snacks and ≥3 snacks per day, as big snackers introduced more calories and were less active. It has been reported that big snackers have higher total energy intake (4,6,8,34). Prospective studies found a direct association between snacking and weight gain risk (3,6–7). Snackers do not compensate for the energy intake excess by reducing caloric intake during the main meals (34). In addition, extra eating occasions are believed to lead to excessive caloric intake because snacks may not generate extra signals of satiety (35). Although a few studies found an increased vigorous activity in snackers (11,13), our results, consistent with most studies (3,5,6), show that those who consumed more or larger snacks were less physically active. Notably, the association between snacking and overweight/obesity remained significant after adjusting for exercise level and total energy intake.
Other studies found an inverse (11,14) or no association (10–13,36) between increased snacking and BMI or weight gain. To justify these findings, it has been hypothesized that splitting foods into multiple eating occasions resulted in reduced caloric intakes at each meal, and that increased eating frequency resulted in a reduction in insulin secretion (37). The reference group for most studies was, however, represented by those who did not consume any snacks (6,9,11) and had a higher BMI, similar to our no-snacking group. An explanation could be that overweight/obese individuals willing to lose weight avoided snacks or tended to underreport their food intakes, particularly snacks. The inverse relation between BMI and meal frequency was no longer significant after excluding adolescents on calorie-restriction diets (38). To avoid the possibility of a systematic bias related to the reverse causality between high BMI and no snacking, we excluded from analyses individuals who had declared to skip snacks. Furthermore, a study evaluating energy expenditure by whole-body calorimetry and doubly labeled water tests did not demonstrate metabolic advantages of multiple eating occasions (38). Because the epidemiologic evidence of the inverse relation between meal frequency and BMI is considered extremely weak, new controlled studies are needed (39). Present conflicting findings may also be the result of the lack of uniformity in the definition of snack (13).
Our most compelling result is the association between timing of the large snack and overweight. Only a few of our participants had the highest-calorie snack in the evening after dinner, and they showed the highest prevalence of overweight/obesity. It is interesting that a higher snacking frequency and the timing of consumption of the largest snacks were both independently associated with overweight/obesity in our participants. The actual trend for adolescents is to eat extremely little in the morning and to shift most daily caloric intake to late in the day (22). When consumed in the afternoon or evening, food seems to induce a weaker sense of satiety, and larger servings tend to be consumed during the day (22). As a consequence, late intakes are more likely to cause positive energy balance. A prospective study showed a significant positive effect of eating later in the day on BMI, suggesting a role of circadian variation in energy intake in the development of overweight (40). These associations were confirmed in a large sample of US children (41). Furthermore, diet-induced thermogenesis is reduced after consuming a snack in the evening compared with the consumption in the morning and afternoon; nocturnal insulin resistance or decreased whole-body fat oxidation has been hypothesized to be responsible for the evening decline of the thermogenic response (20,21). Insulin sensitivity decreases later in the day, and increased levels of free fatty acids and fluctuation in hormonal concentrations may be possible contributors to the circadian modulation of insulin secretion or action (42). In addition, morning gastric emptying may be more rapid than evening gastric emptying (43), and circadian variations in satiety hormones, energy expenditure, and genetic mutations of the circadian clock genes have been associated with weight gain and metabolic abnormalities in humans (44). Finally, the evening snack could be consumed in front of the television or computer, a condition associated with a decrease in consumption of fruit and vegetables and an increase in consumption of sweet beverage and energy-dense foods (3,45).
Adolescence is a period of rapid hormonal variations and growth, leading to profound nutritional changes; the dietary pattern established during this period generally does not change into adulthood, and overweight/obesity during adolescence may increase the risk of adult metabolic diseases. Therefore, it is important to define simple nutritional rules to help maintain healthy BMI. Defining the most convenient quantity, composition, and time of day of scheduled snacks in diets could be an adequate measure to improve adolescent lifestyles.
There is no consensus on what constitutes a snack; our definition was based on a temporal criterion as most studies (8,11,32,33). Indeed, adolescents in our age range are conditioned by their parents in their time of eating. The food-based classification systems may be misleading, because many adolescents consumed sweets, fruit juice, and soft drinks as a part of their meals, and there were no foods/drinks consumed exclusively as snacks. The results did not change after considering as snackers subjects who consumed ≥15% of total energy intake as snacks. This is an observational study, and the possibility of residual confounding cannot be excluded. A 24-hour dietary recall may be inadequate to evaluate the individual typical eating pattern. It is, however, usually used for characterizing habits in groups of individuals (46), also in adolescents (4,47,48). Furthermore, we used the frequency grid as an aid to check the plausibility of answers, and the participants themselves were interviewed by an experienced dietitian, in case of uncertain answers. No considerable discrepancy was found between the measures reported by the parents and the interviewed participants.
We tried to check the accuracy of reporting and analyzed data after excluding under- and overreporters. Nevertheless, we could not exclude that these individuals were accurately reporting their caloric intake (eg, they were overweight and may be trying to lose weight). The prevalence of overweight/obesity did not differ among individuals who ate at the cafeteria and those who did not; most of our participants were <13 years (354/400), and the Centers for Disease Control and Prevention has recently reported that purchasing sodas and snacks at school is much more frequent in individuals >12 years (3-fold higher in 13–15 years and 4-fold higher in 16–18 years) (49). Finally, the comparability of the dietary intakes with the data from national surveys (50) and the strong inverse association between the SDQI score and overweight argued in favor of our data reliability. The study lacked statistical power to analyze the association between the kind of snacks consumed and overweight/obesity. Furthermore, CI coefficients were large, indicating that there was variability in the association between snacking and weight status. Moreover, the majority of children who ate ≥3 snacks or evening snacks were not overweight/obese, thereby suggesting that other protective factors (eg, genetic, environmental, hormonal) need to be studied. Nevertheless, we believe that our results on this controversial matter may be useful, and are worthy of being tested in larger studies.
The strengths of this study were the measurement of all of the anthropometric variables by 2 trained researchers, as opposed to self-reported measures, and the finding that participants in the study were similar to all the adolescents living in Torino with the same age range.
Snacking patterns are associated with overweight/obesity in adolescents. Circadian variation in energy intakes from snacks could be a modifiable factor worthy of being tested in obesity prevention or interventional trials.
1. Larson N, Story M. A review of snacking patterns among children and adolescents
: what are the implications of snacking for weight status? Child Obes
2. Binkin N, Fontana G, Lamberti A, et al. A national survey of the prevalence of childhood overweight
and obesity in Italy. Obes Rev
3. Francis LA, Lee Y, Birch LL. Parental weight status and girls’ television viewing, snacking, and body mass indexes. Obes Res
4. Nicklas TA, Yang SJ, Baranowski T, et al. Eating patterns and obesity in children. The Bogalusa Heart Study. Am J Prev Med
5. Lioret S, Touvier M, Lafay L, et al. Dietary and physical activity patterns in French children are related to overweight
and socioeconomic status. J Nutr
6. Bes-Rastrollo M, Sanchez-Villegas A, Basterra-Gortari FJ, et al. Prospective study of self-reported usual snacking and weight gain in a Mediterranean cohort: the SUN project. Clin Nutr
7. Collison KS, Zaidi MZ, Subhani SN, et al. Sugar-sweetened carbonated beverage consumption correlates with BMI, waist circumference, and poor dietary choices in school children. BMC Publ Health
8. Hampl JS, Heaton CLB, Taylor CA. Snacking patterns influence energy and nutrient intakes but not body mass index. J Hum Nutr Diet
9. Field AE, Austin SB, Gillman MW, et al. Snack food intake does not predict weight change among children and adolescents
. Int J Obesity
10. Phillips SM, Bandini LG, Naumova EN, et al. Energy-dense snack food intake in adolescence: longitudinal relationship to weight and fatness. Obes Res
11. Keast DR, Nicklas TA, O’Neil CE. Snacking is associated with reduced risk of overweight
and reduced abdominal obesity in adolescents
: National Health and Nutrition Examination Survey (NHANES) 1999–2004. Am J Clin Nutr
12. Viskaal-van Dongen M, Kok FJ, de Graaf C. Effects of snack consumption for 8 weeks on energy intake and body weight. Int J Obes
13. Gregori D, Foltran F, Ghidina M, et al. The “snacking child” and its social network: some insights from an Italian survey. Nutr J
14. Ritchie LD. Less frequent eating predicts greater BMI and waist circumference in female adolescents
. Am J Clin Nutr
15. Wansink B, Shimizu M, Brumberg A. Association of nutrient-dense snack combinations with calories and vegetable intake. Pediatrics
16. Hastmann TJ, Bopp M, Fallon EA, et al. Factors influencing the implementation of organized physical activity and fruit and vegetable snacks
in the HOP’N after-school obesity prevention program. J Nutr Educ Behav
17. Drummond S, Crombie N, Kirk T. A critique of the effects of snacking on body weight status. Eur J Clin Nutr
18. Ma Y, Bertone ER, Stanek EJ III, et al. Association between eating patterns and obesity in a free-living US adult population. Am J Epidemiol
19. Jennings A, Cassidy A, van Sluijs EM, et al. Associations between eating frequency, adiposity, diet, and activity in 9–10 year old healthy-weight and centrally obese children. Obesity
20. Romon M, Edme JL, Boulenguez C, et al. Circadian variation of diet-induced thermogenesis. Am J Clin Nutr
21. Hibi M, Masumoto A, Naito Y, et al. Nighttime snacking reduces whole body fat oxidation and increases LDL cholesterol in healthy young women. Am J Physiol Regul Integr Comp Physiol
22. De Castro JM. When, how much and what foods are eaten are related to total daily food intake. Br J Nutr
23. Cacciari E, Milani S, Balsamo A, et al. Italian cross-sectional growth charts for height, weight and BMI (6–20 y). Eur J Clin Nutr
24. Ainsworth BE, Haskell WL, Herrrmann SD, et al. 2011 Compendium of physical activities: a second update of codes and MET values. Med Sci Sports Exerc
25. Pahkala K, Heinonen OJ, Lagström H, et al. Leisure-time physical activity of 13-year-old adolescents
. Scand J Med Sci Sports
26. Bo S, Ciccone G, Durazzo M, et al. Efficacy of antioxidant treatment in reducing resistin serum levels: a randomized study. PLoS Clin Trials
27. Bo S, Bertino E, Trapani A, et al. Magnesium intake, glucose and insulin serum levels in pre-school very-low-birth weight pre-term children. Nutr Metab Cardiov Dis
28. Schofield WN. Predicting basal metabolic rate, new standards and review of previous work. Hum Nutr Clin Nutr
1985; 39 (suppl 1):5–41.
30. Goldberg GR, Black AE, Jebb SA, et al. Critical evaluation of energy intake data using fundamental principles of energy physiology: 1. derivation of cut-off limits to identify under-recording. Eur J Clin Nutr
31. Biltoft-Jensen A, Fagt S, Groth MV, et al. The intake of saturated fat and dietary fiber: a possible indicator of diet quality. Br J Nutr
32. Gatenby SJ. Eating frequency: methodological and dietary aspects. Br J Nutr
33. Summerbell CD, Moody RC, Shanks J, et al. Sources of energy from meals versus snacks
in 220 people in four age groups. Eur J Clin Nutr
34. De Graaf C. Effects of snacks
on energy intake: an evolutionary perspective. Appetite
35. Marmonier C, Chapelot D, Fantino M, et al. Snacks
consumed in a nonhungry state have poor satiating efficiency: influence of snack composition on substrate utilization and hunger. Am J Clin Nutr
36. Maffeis C, Grezzani A, Perrone L, et al. Could the savory taste of snacks
be a further risk factor for overweight
in children? J Pediatr Gastreroentrol Nutr
37. Jenkins DJ, Wolever TM, Vuksan V, et al. Nibbling versus gorging: metabolic advantages of increased meal frequency. N Engl J Med
38. Bellisle F, McDevitt R, Prentice AM. Meal frequency and energy balance. Br J Nutr
39. Mattson MP. The need for controlled studies of the effects of meal frequency on health. Lancet
40. Thompson OM, Ballew C, Resnicow K, et al. Dietary pattern as a predictor of change in BMI z-score among girls. Int J Obes
41. Eng S, Wagstaff DA, Kranz S. Eating late in the evening is associated with childhood obesity in some age groups but not in all children: the relationship between time of consumption and body weight status in U.S. Children. Int J Behav Nutr Phys Act
42. Van Cauter E, Desir D, Decoster C, et al. Nocturnal decrease of glucose tolerance during constant glucose infusion. J Clin Endocrinol Metab
43. Goo RH, Moore JG, Greenberg E, et al. Circadian variation in gastric emptying of meals in humans. Gastroenterology
44. Garaulet M, Gómez-Abellán P, Alburquerque-Bèjar JJ, et al. Timing of food intake predicts weight loss effectiveness. Int J Obes
45. De Craemer M, De Decker E, De Bourdeaudhuij I, et al. Correlates of energy balance-related behaviours in preschool children: systematic review. Obes Rev
46. Thompson FE, Byers T. Dietary assessment resource manual. J Nutr
47. Keast DR, Nicklas TA, O’Neil CE. Snacking is associated with reduced risk of overweight
and reduced abdominal obesity in adolescents
: National Health and Nutrition Examination Survey (NHANES) 1999–2004. Am J Clin Nutr
48. Slining MM, Mathias KC, Popkin BM. Trends in food and beverage source among US children and adolescents
: 1989–2010. J Acad Nutr Diet
49. Mâsse LC, Blanck HM, Valente M, et al. Association between self-reported household practices and body mass index of US children and adolescents
, 2005. Prev Chronic Dis
50. Sebastian RS, Cleveland LE, Goldman JD. Effect of snacking frequency on adolescents
’ dietary intakes and meeting national recommendations. J Adolesc Health